Alphabeta Math
LemmaStatement: AI-generatedProof: AI-generated
How statement and proof provenance work

The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.

  • Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
  • AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
  • AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.

These labels describe origin, not correctness: citations and verification chips remain separate evidence.

Stationary parameter families have a decaying reference-slice residual

Statement

Work in ZFC and fix 0<α<1. Let Σ∗ be a supplied smooth embedded three-dimensional exterior with compact smooth boundary, a fixed end chart x with r=∣x∣>r0, and the fixed compact-core charts defining the global norms Cβk,α(Σ∗). Let (gq,Nq,T) be stationary Lorentzian data on a common neighborhood of Σ∗, jointly smooth in the parameter and ambient coordinates, with one stationary field T and smoothly varying time orientation Nq, for q in an open real interval I0 containing q∗. On a common stationary product end Rt×{x:r>r0} use fixed coordinates (t,x) with T=∂t future timelike, metric coefficients independent of t, and Σ∗: t=u∗(x). The product spatial chart agrees with the fixed end chart, and the complement of this graph end in Σ∗ is compact. Suppose b(q),c(q)>0 are smooth and, on every compact subinterval of I0, for j=0,1 there are common radii and constants giving ∂qj(gq,00+c(q)2)=O5(r−1),∂qj(gq,ij−b(q)2δij)=O5(r−1),∂qjgq,0i=O5(r−2). Write eq,00=gq,00+c(q)2, eq,ij=gq,ij−b(q)2δij and eq,0i=gq,0i. Require also that q↦(eq,00,eq,ij) is C1 in the annular C−14,α norm and q↦eq,0i is C1 in the annular C−24,α norm. These specify the parameter regularity needed here. A sufficient alternative is the same common differentiated bounds also for j=2; no second parameter derivative is otherwise required. Pointwise smoothness in q alone is not substituted for this uniform norm control.

Assume Σ∗ is spacelike and maximal for gq∗, and that its smooth end height satisfies ∣Dju∗∣≤Cr−3/4−j for 0≤j≤4. Then on the end u∗∈C−3/45,α and Δ0u∗∈C−32,α, where Δ0 is the Euclidean Laplacian in x.

There is an open interval I with q∗∈I⊂I0 on which this same reference hypersurface remains spacelike. Choose a fixed admissible transverse flow chart Fw along Σ∗, with F0 its inclusion and with flow field Y=T on a distant end. For each q∈I, let H(q,w) be the mean curvature for gq pulled back by Fw, using the future normal and the sign K(V,W)=gq(∇Vn,W). Thus H(q,0) is a function on the fixed Σ∗, independent of the auxiliary field at w=0. Then q↦H(q,0) is C1 into the global space C−32,α(Σ∗), and H(q∗,0)=0. The neighborhood is local about q∗; spacelikeness on an entire larger parameter interval is not asserted.

The Harrison application uses exactly the mass-one GR/SBS and original maximal-slice hypotheses recorded in the mathematical stationary-primitive and regular-construction suppliers, including their original stationary end, normalized Ernst potential and single chosen smooth normalized primitive W0. For their metric family in the fixed original coordinates, all the stated mixed bounds and parameter-norm conditions hold on every compact J⊂(0,1], with b(q)=q and c(q)=dq/(4q). Around any positive reference parameter with a supplied maximal slice as above, the preceding conclusion therefore applies. At q∗=1 the metric formulas extend smoothly to a common neighborhood of the reference slice for q slightly greater than 1, with the same mixed bounds and a smaller admissible I if necessary. This endpoint extension concerns the metric and its analytic mean-curvature map; no Einstein–Maxwell or electromagnetic-field assertion beyond [0,1] is made.

Statement references

Facts & Assumptions

[F0]

Stationary Lorentzian data fix the signature, smooth time orientation, metric connection and stationary field; an end additionally requires the field to be future timelike. Stationary Lorentzian data and tensor conventions

[A1]

ZFC, including the Axiom of Choice, is carried from the graph bootstrap, transverse-chart and regular-construction suppliers. It is used when invoking those suppliers in steps 1.1, 2.1 and 1.2. The Axiom of Choice

[F1]

The weighted fifth-derivative bootstrap gives u∈C−3/45,α and Δ0u∈C−32,α for its supplied maximal graph in a Minkowski-leading stationary end. The maximal equation upgrades pointwise order-four graph decay to weighted order five

[F2]

Mean curvature is a smooth finite-jet expression on the open set of spacelike graph jets, with bounded derivatives on compact admissible sets. The mean-curvature graph map is a smooth finite-jet expression

[F3]

The differentiated O5 notation and parameter-uniform convention use common Cartesian radii and constants. One extra pointwise spatial derivative gives uniform scaled Hölder control. Differentiated asymptotic bounds and parameter uniformity One extra pointwise derivative supplies scaled Hölder control

[F4]

The flow-graph definition fixes the pullback, future normal and mean-curvature sign. The transverse-chart supplier gives a fixed chart with Y=T on the end once its uniform spacelike and coefficient bounds hold. Normal graphs and the nonlinear mean-curvature map A timelike transverse field gives a fixed-boundary admissible graph neighborhood

[F5]

The global weighted space uses the fixed compact-core norm plus scaled annuli; weighted differentiation, multiplication, positive inversion and smooth coefficient compositions obey the stated uniform Hölder estimates. Weighted Hölder norms on a three-dimensional exterior Weighted differentiation and multiplication estimates Smooth positive inversion and composition in the weighted multiplier class

[F6]

The stationary flux supplier gives the exact normal and mean-curvature flux formula without a Minkowski-leading hypothesis; that hypothesis is needed only for its additional normalized end estimates. Stationary maximal graphs have an explicit uniformly spacelike flux

[F7]

The mathematical primitive supplier, under its full mass-one GR/SBS and original maximality contract, supplies one stationary smooth W0, normalized by W0(T)=d0−D0, whose spatial end components are O5(r−3). The mathematical regular-construction supplier supplies the smooth metric family on the exterior and a common neighborhood of the bifurcation sphere, with f<1 on the exterior and f≤0 on that sphere. Smooth invariant Harrison primitive near the horizon and at infinity Regular Harrison Einstein–Maxwell family

[F8]

The Harrison definitions specify Dq=∣1+q−(1−q)σ∣2, dq=Dq(p) at the supplied pole T(p)=0, the polynomial family Wq=(1−q)2W0−4q(1−q)T♭, and the rank-one metric formula. They also give Wq(T)=dq−Dq. Stationary Harrison one-form and metric Single-potential construction of the Harrison one-form family

[F9]

The mathematical end-denominator supplier uses the seed hypotheses g00=−1+2/r+O6(r−2), g0i=O6(r−2), gij=δij+O6(r−1) and proves σ=1−2/r+O6(r−2) and dq=(1+q)2+(1−q)2ψ(p)2. Its normalized-potential differential and end hypotheses are met by the original end and potential in F7. Mass-one Ernst decay and uniform Harrison denominator bounds

[F10]

Taylor’s integral remainder applies to scalar smooth functions on parameter segments. We apply it componentwise, and to spatial derivatives and two-point differences before taking the weighted suprema. Taylor's theorem with an integral remainder

Proof

Given: The fixed exterior, smooth metric family, mixed bounds and supplied reference maximal slice in the Statement; for the Harrison application use the complete contracts in F7.

Goal: Prove admissibility, the global residual estimate and parameter differentiability, and verify the Harrison mixed bounds and metric endpoint extension.

1.1givenA1F1F3F5F6algebra

At the single fixed parameter q∗ set s=c(q∗)t and y=b(q∗)x. These constant changes make gq∗ Minkowski-leading, with differentiated errors of orders −1,−1,−2 through five derivatives. Its end graph has height v(y)=c(q∗)u∗(y/b(q∗)) and retains all the stated pointwise orders. Under A1 apply F1 in these normalized coordinates; the normalized flux hypotheses follow from the displayed metric bounds and F6. Constant dilations preserve the annular weights, and Δyv=(c(q∗)/b(q∗)2)(Δ0u∗)(y/b(q∗)). Rescaling back gives both asserted end estimates for u∗, in particular p=Du∗∈C−7/44,α.

1.2A1F7F8F9algebra

For the Harrison application use F7 with all its recorded hypotheses, under A1. Its original stationary end and normalized potential meet F9, and its single primitive has the stated five spatial derivative bounds. Write f=−g00 on that end, σ=f+iψ and vi=(Wq)i. The exact formulas in F8 give, for every real q at which Dq≠0, Dq=4q2+eq,vi=(1−q)2(W0)i−4q(1−q)g0i,(Wq)0=dq−Dq. On every bounded parameter interval, finite polynomial differentiation gives ∂qjeq=O5(r−1) and ∂qjvi=O5(r−2) for every fixed finite j, with common constants. Indeed 1+q−(1−q)σ=2q+(1−q)(1−σ), and 1−σ=O6(r−1) by F9; the square modulus is a polynomial in these fixed spatial fields and in q. The seed metric and W0 have no q dependence. No spatial decay is being inferred merely from rational parameter dependence.

2.1givenA1F2F4step 1.1algebra

First choose a compact parameter interval about q∗ inside I0; there b,c are bounded above and away from zero. On the distant end the metric induced on the fixed graph is hq,ij=gq,ij+gq,0iu∗,j+gq,0ju∗,i+gq,00u∗,iu∗,j=b(q)2δij+O(r−1) uniformly in q, since Du∗=O(r−7/4). Thus it has a common positive lower eigenvalue after increasing the end radius. On the complementary compact truncation, hq∗ is positive definite; in finitely many compact coordinate patches its minimum on unit tangent vectors is positive, and joint continuity keeps at least half this bound after shrinking to an interval I about q∗. Bounded coefficients, bounded slope and the displayed tail lower bound are precisely the uniform tail conditions of F4; its compact continuity condition follows from joint smoothness. Apply its chart construction under A1. Hence a single chart can be chosen on a smaller I, and the residual is defined there. This proves admissibility on both the core and the full end.

2.2F5F7F9step 1.2algebra

Fix J=[a,b]⊂(0,1]. Step 1.2 gives Dq≥2a2 on a common sufficiently distant end. On each complementary compact truncation of the reference slice, including its boundary, F7 gives positive Dq on a common ambient neighborhood, so compactness of the parameter/truncation product gives a positive lower bound. For dq use the explicit formula in F9, which is bounded above and away from zero on J. The elementary identity ∂qDq−1=−Dq−2∂qDq, iterated together with spatial differentiation, gives Dq−1=(4q2)−1+O5(r−1) with the same statement for each of the first two parameter derivatives of the error. For example subtract the leading inverse as −eq/(4q2Dq); every differentiated error product still contains an error factor, while its denominators and their inverse derivatives are uniformly bounded at their proper spatial orders. All corresponding core derivatives are bounded by smoothness and the lower bound. These are exactly the quantitative denominator controls needed for mixed estimates.

3.1givenF0F3F5F6F10step 2.1algebra

Write inverse components as gqμν and let Jq=−det⁡gq. On the end define Qq(x,p)=−gq00+2gq0ipi−gqijpipj,Pqi(x,p)=Jq(gqijpj−gqi0)Qq(x,p). F6 gives H(q,0)=Jq−1∂xiPqi(x,Du∗). This uses its exact formula, valid for arbitrary stationary coefficients, rather than its extra Minkowski-leading estimates. The future sign is correct because d(t−u∗)(T)=1 and T is future timelike by F0. On a common small slope ball on a distant end, Qq has a positive lower bound. Inverting the metric gives gq00=−c(q)−2+O5(r−1), gqij=b(q)−2δij+O5(r−1) and gqi0=O5(r−2). For the mixed inverse entry use gqi0=−gqijgq,j0/gq,00. Differentiating the inverse identity, this block identity and the determinant formula gives the same differentiated orders for the first parameter derivative of each error, and Jq=c(q)b(q)3+O5(r−1). Positive denominator bounds and F3, F5 supply their uniform scaled Hölder estimates. Moreover these inverse and density errors are C1 in the corresponding annular C4,α norms: matrix inversion is the cofactor formula divided by a determinant bounded away from zero, and F5 applies to each finite product and smooth scalar composition. Subtracting the constant leading inverse or density leaves at least one original metric-error factor, by the componentwise segment identity in F10; its weight is preserved by F5. The mixed inverse identity preserves its stronger weight −2.

3.2F3F8F9F10step 1.2step 2.2algebra

In the original stationary coordinates, substitute (Wq)0=dq−Dq into the covariant metric formula in F8. It gives the exact component identities (gq)00=−fdq24Dq,(gq)0i=dq4(g0i−fDqvi),(gq)ij=14(Dqgij+g0ivj+g0jvi−fDqvivj). Using steps 1.2 and 2.2 and F9 yields (gq)00=−dq2/(16q2)+O5(r−1), (gq)0i=O5(r−2) and (gq)ij=q2δij+O5(r−1). For the spatial block the last three products are O5(r−4), while (Dq−4q2)gij and 4q2(gij−δij) have order −1. Applying the finite product rule after zero, one or two parameter derivatives preserves each indicated order by steps 1.2 and 2.2. This proves the mixed bounds through j=2 with b(q)=q and c(q)=dq/(4q) on every such J. They also establish the parameter-norm conditions: by F3 the second parameter derivatives of the errors are uniformly bounded in their annular C4,α norms. F10 at orders zero and one gives, in each respective norm, ∥∂qeq+h−∂qeq∥≤C∣h∣ and ∥eq+h−eq−h∂qeq∥≤C∣h∣2. Apply the scalar formulas to spatial derivatives and two-point differences before taking suprema; local smoothness permits the finite parameter integrals. The same argument proves the sufficient alternative in the Statement. Thus the application supplies actual spatial and parameter estimates, without using a physical component-asymptotics item.

4.1F3F5F6F10step 1.1step 3.1algebra

Take x derivatives at fixed p when writing sq(x,p)=Jq−1∑i∂xiPqi(x,p) and put aqij=Jq−1∂pjPqi. Differentiating explicitly gives aqij(x,p)=gqijQq+(gqikpk−gqi0)(gqjℓpℓ−gqj0)Qq3/2. At the constant leading metric and zero slope this is C(q)δij with C(q)=c(q)b(q)−2. The difference at slope p is bounded, with its finite scaled derivatives, by the metric error of order r−1 plus terms of order ∣p∣2+r−2∣p∣+r−4; the absence of a leading linear slope term follows directly from the displayed formula and gi0=0 in the leading metric. Also fq(x)=sq(x,0)=−Jq−1∂xi(Jqgqi0−gq00) has weighted order −3, because gqi0 has order −2. Define Gqj(x,p)=∫01∂pjsq(x,θp) dθ. The fixed-p spatial derivative in sq differentiates a metric coefficient, density or conormal factor; a derivative of a constant leading coefficient vanishes. Thus every term of ∂psq has a spatial metric derivative bounded by O(r−2), with mixed-component derivatives even smaller, and bounded smooth remaining factors. Consequently, after composition with p=Du∗, F3 and F5 give Eqij=aqij(x,Du∗)−C(q)δij∈C−12,α, Gq(x,Du∗)∈C−22,α and fq∈C−32,α on the end. Each of these three coefficient families is C1 in its indicated annular norm. For Eq, subtract its value at the leading constant metric and zero slope by the finite-dimensional segment formula: metric-error factors have weight −1, and the remaining slope factors have weight at most −7/4, hence also include in weight −1. The sharper displayed quadratic slope estimate gives the same bound. For Gq, every term contains a spatial derivative of a metric error, of weight at most −2; its smooth coefficient factors are in weight zero. For fq, expand the displayed derivative: either it differentiates gi0, of weight −3, or it multiplies gi0 of weight −2 by a derivative of a zero-weight coefficient, of weight at most −2. Step 3.1 and F5 now give C1 dependence of these finite products and compositions in their weighted norms. The fixed reference jets are controlled by step 1.1 and have no parameter dependence. The integral identity for sq(x,p)−sq(x,0) follows from F10. Its slope-parameter integrands and their derivatives have the same bounds and a common continuity modulus on 0≤θ≤1; differentiation and two-point differences can therefore be taken before these ordinary scalar integrals and their annular suprema.

5.1F1F2F5F10step 1.1step 2.1step 4.1algebra

The total derivative in the flux formula now gives the precise decomposition H(q,0)=C(q)Δ0u∗+Eqiju∗,ij+Gqj(x,Du∗)u∗,j+fq. The first term is C1 into C−32,α by step 1.1 and smoothness of C(q). The two height-error terms have weights −1−11/4=−15/4 and −2−7/4=−15/4, respectively. By F5 and step 4.1 they are C1 into that weight, which includes continuously into weight −3. The last term is C1 into weight −3 directly. On the compact core, F2 applies on the uniformly spacelike jets from step 2.1; joint smoothness bounds all required spatial and parameter derivatives in finitely many fixed charts. The scalar Taylor formulas in F10, applied also to derivatives and two-point differences, show C1 dependence in the core C2,α norm. Combining the core remainder and derivative-continuity bounds with the uniform annular bounds proves C1 in the global norm: its definition is the sum of the fixed core norm and the annular supremum. Maximality at q∗ gives H(q∗,0)=0.

6.1F7F8F9step 1.2step 2.1step 3.2step 5.1algebra∎

At q=1, D1=4 everywhere and d1=4. On the distant end the fixed seed fields and their recorded derivatives are bounded, so the polynomial estimates of step 1.2 hold also on a bounded interval about 1 and keep Dq positive there after shrinking it. On a compact ambient neighborhood of the remaining reference slice, smoothness and D1=4 give the same positivity on a smaller common interval. The formulas for Wq(T)=dq−Dq and dq=(1+q)2+(1−q)2ψ(p)2 are polynomial identities, so remain valid slightly beyond 1, with dq>0. The rank-one map Aqv=v+Dq−1Wq(v)T has inverse v↦v−dq−1Wq(v)T, and gq=(Dq/4)Aq∗g is Lorentzian there. Thus the rational metric formulas are jointly smooth across the endpoint; step 3.2 still gives their mixed end bounds, while smoothness controls the core. Apply steps 2.1, 3.1, 4.1 and 5.1 about any supplied reference maximal slice, shrinking I for admissibility. This proves the endpoint analytic assertion without differentiating the electromagnetic square-root factor or extending any field equations.

Depends on

Used by · 2

Dependency tree · 2 levels

Each result is drawn at its shortest distance from this item. Arrows run from prerequisites to the items that use them; this item has a heavier outline. External source nodes link directly to the cited text. The view shows up to two dependency steps; open a prerequisite to continue.